Comunicações em Eventos - ODS/03
URI Permanente para esta coleção
A coleção de Comunicações em Eventos compreende trabalhos completos ou resumos de conferências, comunicações orais ou pôsteres, apresentados em congressos, seminários, jornadas, simpósios ou outros tipos de eventos de caráter técnico-científico ou artístico-cultural, publicados em anais impressos ou em meios eletrônicos.
Navegar
Submissões Recentes
- Non-fever COVID-19 Detection by Infrared Imaging(2022) BRIOSCHI, Marcos Leal; NETO, Carlos Dalmaso; TOLEDO, Marcos De; MOREIRA, Mayco Anderson Guedes Maciel; CIVIERO, Nicolas; NEVES, Eduardo Borba; VARGAS, Jose Viriato Coelho; TEIXEIRA, Manoel JacobsenThis study proposed an infrared image-based method for febrile and non-febrile people screening to comply with the society needs for alternative, quick response, and effective methods for COVID-19 contagious people screening. The methodology consisted of: (i) Developing a method based on the face infrared imaging for early COVID-19 detection in people with and without fever; (ii) Recruiting 1206 emergency room (ER) patients to develop an algorithm for general application of the method, and (iii) Testing the method and algorithm effectiveness in 2558 cases (RTqPCR tested for COVID-19) from 227,261 workers evaluations in five different countries. Artificial intelligence was used with a convolutional neural network (CNN) to develop the algorithm that took face infrared images as input and classified the tested individuals into three groups: fever (high risk), non-febrile (medium risk), and without fever (low risk). The results showed that suspicious and confirmed COVID-19 (+) cases characterized by temperatures below the 37.5 degrees C fever threshold were identified. Also, average forehead and eye temperatures greater than 37.5 C were not enough to detect fever similarly to the proposed CNN algorithm. Most RT-qPCR confirmed COVID-19 (+) cases found in the 2558 cases sample (17 cases/89.5%) belonged to the CNN selected non-febrile COVID group. The COVID-19 (+) main risk factor was to be in the non-febrile medium-risk group, compared with age, diabetes, high blood pressure, smoking and others. In sum, the proposed method was shown to be a potentially important new tool for COVID-19 (+) people screening for air travel and public places in general.
- Evaluation of the impact of physical adversarial attacks on deep learning models for classifying covid cases(2022) AGUIAR, Erikson J. de; MARCOMINI, Karem D.; QUIRINO, Felipe A.; GUTIERREZ, Marco A.; TRAINA JR., Caetano; TRAINA, Agma J. M.The SARS-CoV-2 (COVID-19) disease rapidly spread worldwide, thus increasing the need to create new strategies to fight it. Several researchers in different fields have attempted to develop methods to early identifying it and mitigating its effects. The Deep Learning (DL) approach, such as the Convolutional Neural Networks (CNNs), has been increasingly used in COVID-19 diagnoses. These models intend to support decision-making and are doing well to detecting patient status early. Although DL models have good accuracy to support diagnosis, they are vulnerable to Adversarial Attacks. These attacks are new methods to make DL models biased by adding small perturbations on the original image. This paper investigates the impact of Adversarial Attacks on DL models for classifying X-ray images of COVID-19 cases. We focused on the attack Fast Gradient Sign Method (FGSM), which aims to add perturbations to the testing images by combining a perturbation matrix, producing a crafted image. We conduct the experiments analyzing the model's performance attack-free and adding attacks. The following CNNs models were selected: DenseNet201, ResNet-50V2, MobileNetV2, NasNet and VGG16. In the attack-free environment, we reach precision around 99%. When it adds the attack, our results revealed that all models suffer from performance reduction, and the most affected was MobileNet that reduced its ability from 98.61% to 67.73%. However, the VGG16 network showed to be the least affected by the attacks. Our finds describe that DL models for COVID-19 are vulnerable to Adversarial Examples. The FGSM was capable of fooling the model, resulting in a significant reduction in the DL performance.
- Let the data speak: analysing data from multiple health centers of the Sao Paulo metropolitan area for COVID-19 clinical deterioration prediction(2022) VALERIANO, Maria Gabriela; V, Carlos R. Kiffer; HIGINO, Giane; ZANAO, Paloma; BARBOSA, Dulce A.; MOREIRA, Patricia A.; SANTOS, Paulo Caleb J. L.; GRINBAUM, Renato; LORENAT, Ana CarolinaWith the spread of different COVID-19 variants in the Brazilian territory, the national health system has been facing a constant overload. Using data from five different health centers located in the Sao Paulo metropolitan area, this work seeks to identify key common factors associated with the prognosis of COVID-19 severity. The proxies for severity considered are hospitalization time, death and use of mechanical ventilation. The induced models predicted objective short-term COVID-19 clinical deterioration outcomes with AUC, sensitivity and specificity up to 0.880, 0.824 and 0.833, respectively. Parameters such as C-reactive protein and percentage of neutrophils have shown most influence on the predictions. Given the nature of the lab tests highlighted, we note that innate inflammatory status in admission can play a significant role in patient outcome.
- A deep learning approach for COVID-19 screening and localization on Chest X-Ray images(2022) MARCOMINI, Karem Daiane; CARDENAS, Diego Armando Cardona; TRAINA, Agma Juci Machado; KRIEGER, Jose Eduardo; GUTIERREZ, Marco AntonioChest X-ray (CXR) images have a high potential in the monitoring and examination of various lung diseases, including COVID-19. However, the screening of a large number of patients with diagnostic hypothesis for COVID-19 poses a major challenge for physicians. In this paper, we propose a deep learning-based approach that can simultaneously suggest a diagnose and localize lung opacity areas in CXR images. We used a public dataset containing 5, 639 posteroanterior CXR images. Due to unbalanced classes (69.2% of the images are COVID-19 positive), data augmentation was applied only to images belonging to the normal category. We split the dataset into train and test sets with proportional rate at 90:10. To the classification task, we applied 5-fold cross-validation to the training set. The EfficientNetB4 architecture was used to perform this classification. We used a YOLOv5 pre-trained in COCO dataset to the detection task. Evaluations were based on accuracy and area under the ROC curve (AUROC) metrics to the classification task and mean average precision (mAP) to the detection task. The classification task achieved an average accuracy of 0.83 +/- 0.01 (95% CI [0.81, 0.84]) and AUC of 0.88 +/- 0.02 (95% CI [0.85, 0.89]) in 5-fold over the test dataset. The best result was reached in fold 3 (0.84 and 0.89 of accuracy and AUC, respectively). Positive results were evaluated by the opacity detector, which achieved a mAP of 59.51%. Thus, the good performance and rapid diagnostic prediction make the system a promising means to assist radiologists in decision making tasks.
conferenceObject 1 Citação(ões) na Scopus A comparison of three multimodality coronary 3D reconstruction methods(2019) I, Panagiota Tsompou; SIOGKAS, Panagiotis K.; I, Antonis Sakellarios; ANDRIKOS, Ioannis O.; I, Vassiliki Kigka; LEMOS, Pedro A.; MICHALIS, Lampros K.; I, Dimitrios FotiadisThe assessment of the severity of arterial stenoses is of utmost importance in clinical practice. Several image modalities invasive and non-invasive are nowadays available and can be utilized for the 3-dimensional (3D) reconstruction of the arterial geometry. Following our previous study, the present study was conducted to further strengthen the evaluation of three reconstruction methodologies, namely: (i) the Quantitative Coronary Analysis (QCA), (ii) the Virtual Histology Intravascular Ultrasound VH-IVUS-Angiography hybrid method and (iii) the Coronary Computed Tomography Angiography (CCTA). Data from 13 patients were employed to perform a quantitative analysis using specific metrics, such as, the Mean Wall Shear Stress (mWSS), the Minimum Lumen diameter (MLD), the Reference Vessel Diameter (RVD), the Degree of stenosis (DS%), and the Lesion length (LL). A high correlation was observed for the mWSS metric between the three reconstruction methods, especially between the QCA and CCTA (r=0.974, P<0.001).- Development of a System Mobile-based to Assist Asthma Self-Management(2018) SILVA, Thales A.; COSTA, Marly G. F.; STELMACH, Rafael; BLEY, Peter K.; GUTIERREZ, Marco A.; COSTA FILHO, Cicero F. F.Self-management is a major factor in the treatment of asthma and contributes to reduce morbidity in adults and children. However, adherence to self-management depends on a number of factors, including literacy and understanding of disease and health concepts. This paper proposes a system based on the use of mobile devices in order to provide tools to help with adherence to self-management, in addition to propose a narrowing between doctor and patient communication. The proposed system consists of a mobile application for the Android platform, a WEB application and online features of Firebase. In the usability evaluation of the system, most users (82%) rated it as useful and would use the system regularly to support the self-management of asthma.
- A 64-lead Body Surface Potential Mapping System(2017) SALINET, Joao L.; MARQUES, Victor G.; MAZZETTO, Marcelo; CAMARGO, Erick D. L. B.; PASTORE, Carlos A.; CESTARI, Idagene A.Non-invasive acquisition of the electrical heart activity through high density mapping might allow early diagnosis of heart diseases overcoming the limitations of the traditional ECG method. This study presents a BSPM system (hardware and platform) to allow users to analyze the characteristics of morphology in up to 64 simultaneous body surface potentials (BSPs) including the 12-lead ECG and vectocardiogram (VCG). The signals undergo a preprocessing step followed by the R peak detection using previously validated techniques for heart rate variability studies. In addition, embedded 3D isopotential, 3D isochrone maps and VCG planes allow researchers to investigate the heart's the electrical activity and its patterns under different heart rhythm disorders in clinical practice.
- Experimental Method for Recording Epicardium Potentials and Cardiac Myocyte Shortening(2017) MARCHINI, Gustavo S.; TAMASHIRO, Daniel S. U.; OYAMA, Helena T.; CORTELLA, Lucas; CESTARI, Ismar N.; CESTARI, Idagene A.The hemodynamic changes observed in advanced stages of heart diseases are often accompanied by changes in the electrical and mechanical properties of cardiac myocytes. The objective of this work is to develop an experimental method for recording ventricle epicardium potentials in isolated rat hearts and isolated cardiac myocyte shortening. Briefly, rat heart was removed, aorta was cannulated and coronary arteries were retrogradely perfused with heated and oxygenated buffer solutions. Ag-AgCl electrodes fixed in a silicone pouch placed around the heart were used to measure epicardium potentials. The perfusion was switched to an enzyme-containing solution for digestion of the heart and obtaining isolated cardiac myocytes. Measurements of shortening were made in cells electrically stimulated. The results suggest the possibility of relating the electrical behavior of the whole heart with mechanical properties of cardiac myocytes and may represent an useful tool in basic cardiac research.
- Four Years of Experience with the Sao Paulo University Medical School Community Garden(2018) DANTAS, Katia Cristina; ZEMBRUSKI, Paulo Sergio; KUBRUSLY, Marcia Saldanha; CARVALHO-OLIVEIRA, Regiani; MAUAD, ThaisThe School of Medicine, Sao Paulo University Community Garden (FMUSP Community Garden), formed in June 2013, occupies an area of 520 m(2). In the concreted area, vegetables and herbs are grown in large vessels (http://www.facebook.com/HortaDaFmusp). The garden runs on an agroecological basis using locally made compost (garden leaves and horse manure) and bio fertilizers provided by volunteers and the local restaurant (coffee powder). In the garden several herbs, medicinal plants, wild food plants and different types of seasonal vegetables are cultivated. The harvest is open for the entire community. Five medical students received financial support from the University to work 40 h per month to maintain the FMUSP Community Garden. Educational activities for the community include workshops (on medicinal herbs and wild food plants) and cooking events with students and volunteers including an elderly group, focused on healthy eating. In addition, a Ph.D. student conducted studies addressing the role of air pollution on urban gardens using the garden as an experimental site. In summary, the FMUSP Community Garden has provided sustainable, educational and research activities focused on sustainability and healthy eating in the medical campus, on a low budget, for the community. We believe this paper is important because it describes how this experience has benefited many health-related professionals and complements medical teaching. The FMUSP Community Garden has shown that agriculture in large urban centers is possible. The results were very promising, involving students, staff, patients and the surrounding community.
- Application of PMF for Evaluation of the Fine Particles Contribution from Vehicular Emission in Six Brazilian Cities(2014) ANDRADE, Maria; OYAMA, Beatriz; FORNARO, Adalgiza; MIRANDA, Regina; SALDIVA, PauloThe vehicular emission is the main source of fine particles in Brazilian Cities. A comprehensive study was performed from 2007 to 2009 with 24 h daily sampling of fine particles in an experimental site in six Brazilian capitals: Sao Paulo, Rio de Janeiro, Curitiba, Porto Alegre, Recife and Belo Horizonte. The polycarbonate filters collected at each site with Harvard sampling, were submitted to gravimetrical analysis for identification of PM2.5 concentration, to reflectance for Black Carbon concentration, to X-ray fluorescence analysis for elemental composition and to ion chromatography for an ion sand cations composition and concentration. The average PM2.5 concentration were 28, 19, 17, 17, 16 and 11 mu g/m(3) in Sao Paulo, Rio de Janeiro, Belo Horizonte, Curitiba, Porto Alegre and Recife, respectively. Black Carbon accounted for approximately 30 % of the PM2.5 mass concentration in the more air pollution impacted cities: Sao Paulo, Rio de Janeiro and Belo Horizonte. The Black Carbon was used as a tracer for diesel fuel emission and biomass burning. The elemental chemical composition of the PM2.5 was used to identify source-related fractions of fine particles, by means of Receptor Models. The results were used to examine the association of these fractions with daily mortality in each of the six cities. Principal Matrix Factorization (PMF) was applied to the elemental concentration data in order to identify the sources of fine particles, specifically the participation of the vehicular emission. These results were compared to the previous analysis performed with Absolute Principal Component Analysis (APCA). The participation of the vehicular fleet to the PM2.5 mass concentration was significant, explaining in the most urbanized area even 40 % of its mass. These results show the relative importance of the vehicular emission to health injury.